Thank you for acknowledging and trust this issue is given utmost attention and resources for a speedy resolution.
No wind, no snagged cables, no knocking the rig, concrete slabs supporting the tripod and mount-pe200 extension, using 9kgs at 16cm which compensates for the approx 9kg OTA + imagine train + accessories at approx 24 cm. Longitudinally, I balanced the OTA and imagining train on a traditional GE, marked the CG and then aligned/installed on the AM5N motor axis.
A couple of nights ago after ZWO pushed the update, I went outside again and tried removing the weights and tried changing again a range of RA and DEC guiding settings from 500ms to 100ms, 65% to 20%, 1 sec to 0.5 sec, 0.5 and 0.25 Sideral with the same results as shown in pictures attached.





Just to probe the point, I then took the OTA with all the cables and optical train and mounted the whole rig in my 3-year-old iOptron CEM40 EC in the same spot, using the same ASI Air+ where moments before the AM5N had been tested.
I have attached a couple of subs with narrow band filters of the Lion’s Nebula, 300 secs, -10C and the performance over 200 secs of the optical setup using the iOptron mount. I believe this is fairly strong suggestion that there is something wrong with the AM5N, however, I am willing to be convinced otherwise.
Tried NINA and PHD2 and I continued to experienced the same departures in the RA as with the ASI Air+ controlling the rig. I did get so wound up after the 2 departure in 5 minutes (off the 8" scale) I took all the rig in and reconnected the ASI Air+ to my CEM40 EC and got some lovely subs of the Flying Bat Nebula as consolation.